I also think the economic side of the game is incredibly difficult to penetrate-- and, even when one does figure out how it works, runs counter-intuitive to how you think it should. I think the game would be far better suited to abstracting most of the economics rather than trying (and ultimately failing) to realistically simulate supply and demand. Abstraction doesn't have to mean simple, it's just that a few game issues really need to be confronted:
* Shortages of goods due to reasons you as a player cannot figure out, and likely due to the AI not knowing how to operate its own economy.
* Economic ebb and flow, all of it presenting more information that you can possibly disseminate and have only indirect control over.
So ignore supply and demand. Let every good have a set "value", and if that value needs fluctuation let it relate to things that would normally fluctuate value like wars and such. Base a country's economy on how much of a good the people have access to... so do you produce it yourself? If not, do you have good trading infrastructure? Are you being blockaded, or do you have trade routes overland with railroads? Do you have trading partners (being able to arrange trade deals as a diplomacy option would be nice)? Do you have tarriffs, ones that actually protect local industry at the expense of making access to foreign goods harder? All of this could contribute to "ease of access" your people/industries have to goods and THAT should be the determining factor for how well your economy does. Something that the player can actually do something about.
I think we all learned a lot from V2, and that looking back there's a lot of stuff about the economy the devs would do differently. I know a couple of dozen ways you could code round the economic problems of V2, but only because I've been able to look at V2 to develop from - so I don't really think there's any need to do away with price fluctuations or supply and demand. Some of the market mechanics need addressing, tho.
First up, I'd ditch buying order on the WM. Everyone buying from WM buys simultaneously.
An SOI owner would buy first from their sphere's market, but after them all other members of the sphere buy simultaneously.
Every continent would have a 'regional market' SOI, which was just a passive SOI with no leader (so everyone in it buys at the same time). They'd buy from the home continent before the WM, but after their internal market.
If a POP works in a RGO that produces a Life Need, it takes it's share of that need in lue of cash.
I'd also have money work entirely differently, since I think money's current design in the game is fundamentally flawed from the first concept; it was approached from the point of view of an in-game resource, like in EU3, but in Vicky money needs to operate as a medium of exchange. Money always has to come from something and go somewhere when it's spent, rather than vanishing in a puff of smoke like it can in the other games. This is a major, pretty unsolvable problem with the V2 model.
You need a working debt mechanism which the pops can use, and you need leverage and an interest rate. You need an elastic and responsive inflation device, which precious metal mines are completely insufficient for. And you need a means to distribute the money.
So i'd use banks. I'd create the banking sector in-game, and I would have each bank store a set amount of money, and be allowed to lend upto X times that amount of money to POPs at interest (set by government policy, or by a slider in the budget panel). The bank would get money back from the interest, allowing it to lend out more money. This would permit a flexible cash supply - if there was too little money in the system, POPs will borrow and spend, and if there's too much money, POPs will earn too much and pay off their loans, preventing the cash from growing (and reducing the amount in circulation as the cash flows back to the bank). If POPs fail to pay their interest, the bank takes the hit, but if POPs are earning cash, the bank swoops in before they can spend any (note that this excludes farmers eating their product straight from the field).
Sovereign debt, I would switch to a bond issuing system, where a country issues bonds at their own interest rate. Banks will automatically buy bonds issued at interest rates equal to or higher than their own (each bond being bought in equal parts by all eligible banks). Countries may also buy bonds in each other. If you're flat broke and a bond fails to sell, then a country goes bankrupt. These mechanics should automatically produce a credit system which forces the player to bid increasingly high interest rates when trying to sell bonds, but a simple credit system using event mods is perfectly possible.
So, with that we'd a) be able to have financial powerhouses like Switzerland, b) have solved the cash capping the WM issue, c) have solved the debt making no sense problem, and d) have produced an interesting, believable and reasonably straightforward banking system.
